• Since when is 2,546 miles shorter than 1,000 miles?

• Image scaling!  All I can see is that you are asking directions and there are are suggested routes.

• For me the scale is the same whether it's the last or last-1 zoom level, probably a simple mistake.

• @locallunatic said:

Image scaling!  All I can see is that you are asking directions and there are are suggested routes.
Sorry, better?

Looks like Google Maps is using a Mercator projection and the scale is only accurate for the vertical center of the map.  Too bad they didn't actually center the scale vertically on the map.

• @Pascal said:

Looks like Google Maps is using a Mercator projection and the scale is only accurate for the vertical center of the map.  Too bad they didn't actually center the scale vertically on the map.

Nope, the scale changes as you move vertically. It seems (as i stated above) that they simply forget to multiply the zoom factor at the last zoom level (or use the factor from last level-1).

•  @swayde said:

Nope, the scale changes as you move vertically. It seems (as i stated above) that they simply forget to multiply the zoom factor at the last zoom level (or use the factor from last level-1).
Looks to me like the scale changes as you move vertically such that it is accurate for the vertical center of the map.  When I place the route it mapped out in the vertical center of the map and compare the distance to the scale it looks about accurate.

• @Pascal said:

Sorry, better?

Much, thanks.

• @Pascal said:

@swayde said:

Nope, the scale changes as you move vertically. It seems (as i stated above) that they simply forget to multiply the zoom factor at the last zoom level (or use the factor from last level-1).
Looks to me like the scale changes as you move vertically such that it is accurate for the vertical center of the map.  When I place the route it mapped out in the vertical center of the map and compare the distance to the scale it looks about accurate.

This.  Which is about as sensible as they could have done.  The route will likely be in the center of the screen.

• @Sutherlands said:

@Pascal said:
Looks to me like the scale changes as you move vertically such that it is accurate for the vertical center of the map.  When I place the route it mapped out in the vertical center of the map and compare the distance to the scale it looks about accurate.
This.  Which is about as sensible as they could have done.  The route will likely be in the center of the screen.
Silly me, I would have used a Goode homolosine projection or put the scale on the map at a location where it was actually accurate.

• @Pascal said:

@Sutherlands said:

@Pascal said:
Looks to me like the scale changes as you move vertically such that it is accurate for the vertical center of the map.  When I place the route it mapped out in the vertical center of the map and compare the distance to the scale it looks about accurate.
This.  Which is about as sensible as they could have done.  The route will likely be in the center of the screen.
Silly me, I would have used a Goode homolosine projection or put the scale on the map at a location where it was actually accurate.

Which is why nobody uses Pascal maps, they use google maps.

• @Pascal said:

I would have used a Goode homolosine projection

Probably not a goode idea for a seamless draggable map:

This grafting results in a kink in the meridians along the parallel of the graft.

But I agree with the sentiment which is that Mercator needs to be fucking retired already.

• @dhromed said:

But I agree with the sentiment which is that Mercator needs to be fucking retired already.

TMSO or GTFO!

• @Pascal said:

Silly me, I would have used a Goode homolosine projection or put the scale on the map at a location where it was actually accurate.

I'd put it in a simple polar projection, so that nobody mistakes the borders of the map for something usefull...

But well, I'd do that after I think about the issue, and as absurd as it sounds, it is quite possible that one puts a huge maps service out there and don't stops even once to think hard about projections.

•  Fractals!

• Breaking news!

The world isn't flat!

• @Pascal said:

Silly me, I would have used a Goode homolosine projection

That's about as useless as it can get. For the map it's most important that at large scale (local view) it is minimally distorted. And Mercator is the projection that matches local azimuthal projection at large scale and scales smoothly to the whole world (well, it can't show anything beyond ±85°, but fortunately there is not much to map in the polar regions).
@Pascal said:
or put the scale on the map at a location where it was actually accurate.

… or not put it on the map at all when the scale differs by more than few percent across the map.

• @Bulb said:

Mercator is the projection that matches local azimuthal projection at large scale

Yes, but the downside is that Greenland becomes as big as Africa.

• @Mcoder said:

I'd put it in a simple polar projection
This.

Bonus points if you could freely rotate the map around its central point, the initial angle being given by compass information when available.

PS: wouldn't be nice if we could then load these maps in a portable device? I'd call it the "universal locationing system".

• @dhromed said:

@Bulb said:

Mercator is the projection that matches local azimuthal projection at large scale

Yes, but the downside is that Greenland becomes as big as Africa.

It simply doesn't matter when 99.9999% of the use is going to be on a local scale.

• @Sutherlands said:

@dhromed said:

@Bulb said:

Mercator is the projection that matches local azimuthal projection at large scale

Yes, but the downside is that Greenland becomes as big as Africa.

It simply doesn't matter when 99.9999% of the use is going to be on a local scale.

But even then it still fails.

• @dhromed said:

@Bulb said:

Mercator is the projection that matches local azimuthal projection at large scale

Yes, but the downside is that Greenland becomes as big as Africa.

Clearly they should use a Dymaxion projection instead!

• @Renan said:

@Sutherlands said:
@dhromed said:

@Bulb said:

Mercator is the projection that matches local azimuthal projection at large scale

Yes, but the downside is that Greenland becomes as big as Africa.

It simply doesn't matter when 99.9999% of the use is going to be on a local scale.

But even then it still fails.

That has absolutely nothing to do with this thread.

• @Sutherlands said:

That has absolutely nothing to do with this thread.

• @boomzilla said:

@Sutherlands said:
That has absolutely nothing to do with this thread.

Since when is Roy Orbison Asian?

• @boomzilla said:

@Sutherlands said:
That has absolutely nothing to do with this thread.

I think that's related to google maps.  Yes, you have it correct.

• @boomzilla said:

@Sutherlands said:
That has absolutely nothing to do with this thread.

Supreme Commander dictate that Motherland must be largest country on map! Get projecting!

• @da Doctah said:

@boomzilla said:

@Sutherlands said:
That has absolutely nothing to do with this thread.